The effect of data sampling time on the accuracy of maximum output power of a thermoelectric module

Abstract

Abstract The instantaneous short-circuit current and open-circuit voltage of TEM (Thermoelectric Module) can be used to quickly estimate its maximum output power (P omax). Because of their instantaneous samples, the sampling time has a great influence on the accuracy of the estimated result. In this paper, the influence mechanism of sampling time is analysed systematically and a module of TEG1-127-2.8-1.2-250 is taken as a sample for verification. By means of either steady open-circuit/instantaneous short-circuit at mode A or steady short-circuit/instantaneous open-circuit at mode B, P omax can be estimated quickly. At single mode A or B, the shorter the sampling time is, the greater the relative error of the estimation of P omax has. This error varies with different temperature of hot plate, but the variation trend with sampling time is consistent. The valuations at model A and B are larger and smaller than the accurate values, respectively. When the sampling time is 600 to 1000ms, the error is the smallest (less than 3%). The accuracy of P omax is significantly improved by using the average results of mode A and B. In this way, the sampling time has little influence on the estimation error, and the shortest sampling time corresponds to the highest estimation accuracy.